One of the more obscure provisions in the nuclear power subtitle of the discussion draft of the “American Power Act” (or simply “K-L”) circulated by Senators Kerry and Lieberman last week is the blandly titled Section 1108, “Inspections, Tests, Analyses, and Acceptance Criteria.” This section, comprising a mere six lines of text, would strike out a sentence of the Atomic Energy Act and replace it with the following: “Following issuance of the combined license, the [Nuclear Regulatory] Commission shall ensure that the prescribed inspections, tests, and analyses have been met.”
On first reading, this looks innocuous enough —- after all, what could be wrong with requiring that the NRC make sure that all inspections, tests and analyses prescribed in a nuclear plant license have been met? But comparison of the suggested revision with the original text should suggest even to the casual reader that there is something fishy going on. The current language in Section 185 b of the AEA says that “Following issuance of the combined license, the Commission shall ensure that the prescribed inspections, tests, and analyses are performed and, prior to operation of the facility, shall find that the prescribed acceptance criteria are met.” So the K-L revision would do two things: remove the phrase “prior to operation of the facility” and change the tense of the verb so that “are met” becomes “have been met.”
Those who are not steeped in the Byzantine world of nuclear plant licensing may be forgiven for not understanding the major significance of this seemingly minor revision. In fact, this change would settle, by congressional fiat, a years-long dispute between the NRC and the nuclear industry to define the scope of the information that the NRC needs before formally deciding that a newly constructed nuclear plant is safe to operate. And it would resolve the dispute in the industry’s favor.
The heart of the issue is this: for a new nuclear plant there will be as many as 900 “inspections, tests and analyses” that must be performed to demonstrate that the plant has been built in accordance with design specifications. Some of these can be performed relatively early in the construction process. But it may take many years for the plant to be completed. How can the NRC make sure that an inspection, test or analysis done five years ago remains valid when the plant is finally ready to operate, especially with heavy construction taking place all around the piece of equipment in question?
The current language in the AEA clearly states that the NRC must make sure that all acceptance criteria are met —- that is, in the present tense —- before it can permit the facility to operate. Taken literally, this would mean that the NRC’s startup authorization must await a “perfect day” in the life of the nuclear plant, when it knows that every system, structure and component is installed properly and has been maintained so that it will work as designed. But the industry has complained that perfection is not attainable and pointed out that the NRC allows current plants to operate even when they are not in perfect condition or when systems are out of service. The industry believes that the NRC should not hold new plants to a different standard.
In fact, the NRC agrees with the industry that “the perfect day” is not realistic, and the remaining differences in the public positions of the two sides are small. But the NRC also believes that the current regulations do not provide it with enough tools to determine when an “imperfect day” is good enough for the plant to begin operation. So therefore, over industry objections, it has begun the process to develop new regulations governing how nuclear plant licensees should maintain the knowledge base for systems, structures and components that have already passed their inspection criteria throughout the entire construction period.
Here’s where K-L comes in. By changing the law so that “are met” becomes “have been met,” it would undermine the regulatory basis for the new rules. For if the acceptance criteria for a particular component have been met at one time, the NRC would not have the authority to require that they continue to be met during construction. And even worse, the removal of the phrase “prior to operation of the facility” would also eliminate the NRC’s legal authority to make its startup authorization contingent on a finding that all acceptance criteria “have been met.” So if the K-L substitute language becomes law, it appears that the NRC could not block startup of a new nuclear plant even if critical safety components have never been tested. This would severely limit the NRC’s ability to prevent startup of a new nuclear plant with serious defects. Little wonder that this wording change was one of the legislative priorities of the Nuclear Energy Institute, the main lobbying arm of the nuclear industry.
To be sure, we do not believe that it is the industry’s desire to start operating new plants without taking steps to ensure that critical safety systems will work. But plant licensees may have less stringent criteria than those of the NRC for deciding when a system will “work,” especially if they are under economic pressure to get a new plant up and running. This is why the NRC needs the authority to ensure that no new plant should start up until a day that, if not perfect, is still pretty darned wonderful.
UCS does not believe that it is appropriate for Congress to waive a magic wand and short-circuit the NRC’s deliberative process, especially when it is dealing with subtle and heavily technical issues. NRC should be allowed to follow its current path for resolving these difficult questions through rulemaking, which will afford both industry and the public with the opportunity to review and comment on the proposed rules. Senators Kerry and Lieberman should remove Section 1108 before their important climate and energy bill is finalized.
Support from UCS members make work like this possible. Will you join us? Help UCS advance independent science for a healthy environment and a safer world.